johko
/

capdec_015

johko commited on Dec 30, 2022

Commit

319dcd9

1 Parent(s): bc2330d

Update README.md

Files changed (1) hide show

README.md CHANGED Viewed

@@ -6,7 +6,7 @@ license: apache-2.0
 This is are model weights originally provided by the authors of the paper [Text-Only Training for Image Captioning using Noise-Injected CLIP](https://arxiv.org/pdf/2211.00575.pdf).
-There method aims to train fine-tune CLIP with only text samples. Therefore they are injecting zero-mean Gaussian Noise with a standard-deviation(STD) of into the text embeddings before decoding.
 In their words:
 *Specifically, we assume that the visual embedding corresponding to a text embedding

 This is are model weights originally provided by the authors of the paper [Text-Only Training for Image Captioning using Noise-Injected CLIP](https://arxiv.org/pdf/2211.00575.pdf).
+Their method aims to train CLIP with only text samples. Therefore they are injecting zero-mean Gaussian Noise with a standard-deviation(STD) of into the text embeddings before decoding.
 In their words:
 *Specifically, we assume that the visual embedding corresponding to a text embedding